If the lifetime contribution of one client is a large multiple of what it costs you to win one, keeping clients is cheaper than winning new ones. In the worked example below, keeping a client for five years costs £688 and winning their replacement costs £1,789.
When the numbers get tight, owners often reach for more leads: a bigger ad budget, a rebuilt website, a lead generation agency on a three-month trial. Meanwhile clients may be leaving at a rate nobody has measured.
Why does the leak stay invisible?
Clients rarely leave with a complaint. They stop ordering or order less, or the person who chose you moves on and their replacement brings a supplier of their own. A lost client produces no event, no date and no number, so it never appears on the week’s list.
The four calculations below take about twenty minutes with your sales ledger.
Step one: what is your churn rate?
Count the clients who stopped buying from you over the last full twelve months, then divide that by the number of active clients you had at the start of those twelve months. That percentage is your annual churn rate.
Two rules keep the number accurate. Use a full year, not a quarter, or a seasonal business will read a quiet spring as collapse and a busy autumn as loyalty. And define “active” once, in writing, before you count: a client who has not ordered in eighteen months has already left, whatever the CRM says. Lapsed clients left on the list make churn look lower than it is.
Step two: how long does the average client stay?
Divide one by your churn rate. That is your average client lifetime in years. Lose a quarter of your clients each year and the average relationship lasts four years. Lose a tenth and it lasts ten. Because the relationship is not linear, a small fall in churn adds years to the average relationship.
One caveat: this holds when churn is roughly steady across the base. If most of your leavers go in their first six months, the formula flatters you, because it spreads early departures across clients who were never at risk. If that is your pattern, run the calculation twice — once for clients under a year old, once for the rest — and you may find an onboarding problem rather than a retention problem, which is a different fix.
Step three: what is a client worth?
Value a client by what they contribute over that lifetime.
Take an illustrative design and print studio: 48 active clients at the start of the year, each ordering roughly six times a year at an average order value of £850. That is £5,100 of revenue per client per year and £244,800 in total, all figures excluding VAT — this business is well over the £90,000 VAT registration threshold that has applied since April 2024. After direct costs the studio keeps 45p in every pound, so each client contributes £2,295 a year.
Twelve of those 48 clients stopped ordering during the year. That is a churn rate of 25%, so the average client lasts four years, and the average client is therefore worth £20,400 of revenue and £9,180 of contribution over the life of the relationship. Corporation tax takes its share of whatever reaches taxable profit: for the financial year beginning 1 April 2026 the small profits rate is 19% on profits up to £50,000, rising through marginal relief to the 25% main rate above £250,000, so at this studio’s profit level £9,180 of contribution is around £7,436 kept.
Every client who leaves takes roughly £9,180 of contribution with them, and the studio lost twelve of them last year without a conversation.
Step four: what does a replacement cost?
Add up everything you spent last year on winning new clients, then divide by the number you won. Everything means advertising, the website work, the directory listings, any commission — and the selling time, costed properly, because that time is often the largest line and the one left out.
The studio spent £16,800 on marketing over the year. It quoted 60 jobs for new prospects at about two and a half hours each, which is 150 hours; costed at £55 an hour of owner and estimator time, that is £8,250. Total £25,050, and it won 14 new clients. So each new client cost £1,789 to acquire.
Now price a contact programme. Two structured 45-minute conversations a year with every client, plus a monthly note that takes five minutes to personalise, comes to roughly two and a half hours per client per year. At the same £55 an hour, that is £137.50 per client per year, or £6,600 across the whole base.
So which is cheaper — keeping or winning?
Suppose that programme moves churn from 25% to 20%, so one client in twenty who would have drifted away stays. Average lifetime goes from four years to five. Lifetime contribution per client goes from £9,180 to £11,475, an extra £2,295, and the contact time that produced it costs about £688 across the whole five years.
Run it as an annual flow instead and the same thing shows up. At 20% churn the studio loses 9.6 clients a year rather than 12. Those 2.4 retained clients keep £5,508 of contribution in the business and save £4,294 of acquisition cost that no longer has to be spent replacing them — £9,802 of benefit against £6,600 of cost, so £3,202 better off in year one, and better again every year after, because the base it compounds on is larger.
What does the Five Ways lens show?
The Five Ways chain is simple: leads, multiplied by conversion rate, gives customers; customers, multiplied by the number of transactions and the average value of each, gives revenue; revenue, multiplied by margin, gives profit. Five levers, and improving any one of them improves profit.
Lead generation moves the first lever, and it has to be bought again every month.
Retention moves two, and moves them on the base you already have. It raises the number of transactions, because a client who stays five years buys thirty times rather than twenty-four. It raises average value, because established clients take on the second service, ask for the bigger version and stop shopping the price.
Compare the two. A fifth more leads takes the studio from 140 enquiries to 168, and at a 10% conversion rate that is 2.8 extra clients worth £6,426 of contribution. It costs a fifth more marketing spend (£3,360) and a fifth more quoting time — twelve more quotes at two and a half hours, £1,650 — so £5,010 all in, for a net gain of £1,416. The retention programme returned £3,202 for the same year.
What is the decision rule?
Before the next spend, make one comparison: what would a fifth off your churn rate be worth, and what would a fifth more leads be worth?
Retention wins whenever your lifetime contribution per client is a large multiple of your acquisition cost — the studio’s is £9,180 against £1,789, a ratio of 5.1 to 1 — because at that ratio, keeping is buying the same contribution at a fraction of the price. If your ratio is under about 3 to 1, you have a pricing or a margin problem underneath the retention problem, and our guide on pricing for profit is the better place to start. A base that leaks and a margin that is too thin can both cause a plateau, and why most small businesses plateau covers how to tell which one is binding.
Why does this get ignored?
Winning a client is an event with a date, a name and a number, and it gets announced to the team. Keeping a client produces nothing to point at in the Monday meeting.
So growth gets measured in wins, and leavers are absorbed one at a time, each with its own explanation. In the example, twelve explanations add up to a 25% churn rate that nobody owns.
Fix it the way any other invisible outcome gets fixed: by giving it a named owner, a measure and a fixed rhythm.
What to do this week
Do the four calculations in one sitting.
- Count leavers over the last twelve months, divide by the client count at the start, and write down your churn rate.
- Divide one by that rate. That is your average client lifetime in years.
- Multiply your average annual revenue per client by your contribution margin, then by that lifetime. That is what a client is worth.
- Add last year’s total marketing spend to your costed selling time and divide by new clients won. That is what a replacement costs.
Then put three rhythms in the diary and treat them as fixed.
- A quarterly at-risk review. Ninety minutes, whole client list, scored against signals rather than instinct: spend down more than 20% year on year, no contact initiated by them in 90 days, the person who chose you has left, a complaint fixed but never followed up, or a sudden request for a price breakdown. Two or more signals means somebody calls them this month.
- A defined re-contact cadence. Written down, by client tier, with a name against each. For example, quarterly for the top tier and twice a year for the rest, with a named person accountable for it happening.
- An exit conversation with every leaver. Thirty days after they go, not on the day, once the defensiveness on both sides has dropped. Three questions: when did you decide, what would have changed it, and what did the alternative offer that we did not? Log the answers in one place. After six, look for the pattern.
A monthly business review keeps this arithmetic in view, and the first quarter of business coaching often covers it. If holding the line on price is the part you expect to be hardest, our guide on difficult client conversations is the place to start.
Get the next one in your inbox
One practical guide for business owners each week. Unsubscribe at any time.
Frequently asked questions
How do I work out my churn rate if my client list changes constantly?
Count leavers over a full twelve months and divide by the number of active clients you had at the start of those twelve months. A client who left counts once, on the date of their last invoice, regardless of when they joined. Use a full year rather than a quarter, because a seasonal business will read a quiet spring as collapse and a busy autumn as loyalty. If your client base grew sharply during the year, run the calculation on the clients who were there at the start only, so you are measuring what happened to a fixed group rather than diluting your leavers across a base that did not exist for most of the period.
Isn’t raising prices a faster route to the same profit?
It is faster, but the order matters. A price rise applied to a leaking base accelerates the leak, and the clients who go first are a mix of the price-sensitive ones you were quietly subsidising, which is fine, and the loyal ones who felt taken for granted, which is expensive. Fix the contact rhythm first and the same increase lands differently, because a client who has had two proper conversations with you this year reads it as a business decision rather than an opportunistic grab. Do both, in that sequence, roughly a quarter apart.
What if the clients I’m losing are ones I didn’t want anyway?
Check that rather than assume it. Split the leavers into two lists: those you would take back tomorrow and those you would not. Deliberate churn of unprofitable clients is portfolio management and belongs in your margin story, not your retention story. Look at the ratio. If ten of the twelve are clients you wanted, the arithmetic above applies to those ten. For the clients you would not take back, record why, and check whether the reason was price, scope or fit, because that tells you which new clients to turn away at the enquiry stage.
Does keeping clients mean discounting to hold on to them?
No. A discount offered at the point of leaving often buys only a few months and teaches every client who hears about it that your price is negotiable under pressure. The retention work here is contact: two structured conversations a year, a defined cadence and a quarterly at-risk review. If a client can only be kept by cutting the price, the relationship was already transactional, so let them go and record the reason. Record the reason in the same place as your exit conversations.